UC San Diego

Arctic sea ice retreat is a well-known signatures of climate change and is driving changes to the underlying ocean. Progress in modeling and observational systems have driven huge advances towards better understanding the complexities related to Arctic sea ice retreat and Arctic Ocean processes. The objective of this thesis is to incorporate the results of these studies into the development of simple but informative methods aimed at increasing our understanding of simulated sea ice evolution in climate models and ocean processes in observations.

Chapters 2 and 3 are focused on examining sea ice changes in climate models with respect to simulated levels of global warming. In contrast to many previous studies that did not consider simulated global warming, we find that simulated internal variability cannot explain differences between observed and modeled sea ice retreat (Chapter 2). Next, we examine sea ice retreat in the two most recent generations of climate models and find that previously reported improvements in the simulated sea ice retreat was caused by an increase in the global warming bias, driven by the inclusion of simulated volcanic forcing (Chapter 3).

Chapters 4 and 5 are focused on examining changes in the mixed-layer evolution during the melt season in the Canada Basin in 1975, 2006, and 2007. Using a simplified salt budget, we find that recent increases in the seasonal mixed-layer freshening could mainly be a result of shallower mixed layers, which act to concentrate freshwater input within a smaller volume (Chapter 4). Motivated by this, we use a simplified energy budget to investigate factors driving differences in the mixed-layer depth evolution during the melt season in these three years. The results suggest that reduced ice-ocean drag may play a significant role in explaining shallower mixed layers in recent years (Chapter 5).